Proton-detected 2D heteronuclear shift correlation via multiple

Jurgen M. SCHMIDT , Harald THURING , Angela WERNER , Heinz RUTERJANS , Rainer QUAAS , Ulrich HAHN. European Journal of Biochemistry 1991 197 ...
0 downloads 0 Views 273KB Size
J . Am. Chem. SOC.1990, 112, 1279-1280

1279

to obtain noticeable concentrations of Ph3C+. Conclusion. It has been demonstrated that, in a polar solvent such as water, a neutral radical can be photoionized to give the corresponding carbocation. In most cases, the rate of this conversion is likely to be limited only by the length of the ionizing light pulse, in which case it should be possible to produce and study ultrareactive cations with lifetimes, e.g., in the picosecond domain. Acknowledgment. J.L.F. thanks the Deutscher Akademischer Ausstauschdienst (DAAD) for a stipend.

X

1 8 O X 9OP

QOX*TPPI

Dec

t,

Proton-Detected 2D Heteronuclear Shift Correlation via Multiple-Quantum Coherences of the Type 12S

GARP

Figure 1. Pulse scheme for heteronuclear shift correlation via 12S three-quantum coherences. The delay A should match l/(WNH). The phases are cycled as follows: 8 = x, -x; = x, x, -x, -x, y, y , -y, -y; ref = x , -x, -x, x , -y, y, y, -y.

+

Jurgen M . Schmidt and Heinz Ruterjans* Institut fur Biophysikalische Chemie Johann Wolfgang Goethe- Uniuersitiit Theodor-Stern- Kai 7, Haus 75A 0-6000Frankfurt am Main 70, FRG Received August 23, 1989 NMR spectroscopy of high molecular weight compounds suffers from overlap problems giving rise to severe difficulties in spectral assignments. 3D NMR measurements' are helpful in simplifying spectral patterns but require an enormous amount of computational effort. Alternatively, filtering methods offer the possibility of selectively extracting spectral information from isotope-labeled compounds.2 In order to provide maximum sensitivity, these so-called editing techniques very often are based on the application of polarization transfer protocols in conjunction with proton detection. We are reporting here on such a pulse sequence, which allows one to exclusively detect heteronuclear spin systems of the type 12S,I and S representing protons and heteronuclei, respectively. This type of selection is extremely useful in discriminating asparagine and glutamine side-chain amide protons from those located in the peptide backbone, as will be shown for uniformly ISN-labeled RNase Ti. Although the NH, resonances could be identified on the basis of the 0.6 ppm N H D isotope effect, they are likely to be covered by other signals, especially in spectra of larger proteins. In the following, the basic ideas leading to the design of this particular 2D N M R experiment are summarized. The large heteronuclear one-bond coupling of approximately 90 Hz allows fast excitation of proton-heteronucleus multiple-quantum coherences, including the nitrogen nucleus as a relay site. Usually, the homonuclear coupling constants are 1 order of magnitude smaller and require much longer preparation periods. The selection between different spin topologies depends on these correlations. Consequently, resonances from side-chain amides can be distinguished from those of backbone amides. A modified DEPT sequence has proven to be suitable for the stepwise preparation of triple-quantum c o h e r e n c e ~ . ~ ?From ~ symmetry considerations, we concluded that a time-reversed DEPT-type sequence would allow the conversion of the triplequantum coherences into detectable single-quantum terms. This leads to the pulse sequence depicted in Fwre 1. In a back-tc-back manner, both DEPT pulse clusters flank the evolution period during which chemical-shift labeling of protons with the heteronuclei precession frequencies takes place. Proton chemical-shift contributions were removed by employing selective inversion of proton populations in the middle of the evolution period. Thus, triple-quantum and single-quantum coherences were exchanged, (1) Griesinger, C.; Ssrensen, 0. W.; Ernst, R.R.J. Mogn. Reson. 1987, 73, 514-519. ( 2 ) Torchia, D. A.; Sparks, S. W.; Bax, A. Biochemistry 1989, 28, 5509-5524. ( 3 ) Dodrell, D. M.; Pegg, D. T.; Bendall, M. R.J . Mum.Reson. 1982,48, 323-327. ( 4 ) Ssrensen, 0. W.; Jakobsen, H. J. In Pulse Merhods in 1D ond 2D Liquid-Phase NMR; W. S. Brey: San Diego, 1988; pp 252-253.

0002-7863/90/1512-1279$02.50/0

i

~

,(I

v

F2

-

r

0

,

8

,

7

,

I

6

116

It

ppm

Figure 2. Selective heteronuclear shift correlation spectra for terminal amide groups in side chains of asparagine and glutamine residues (A) and for all amides (B) in uniformly 15N-labeled RNase TI; the I5NH2 correlations are indicated in the expansion (C). Both spectra are repthe 256 increments of resented in a pure absorption mode. Prior to FT, 2K data points were zero-filled once in t2 and twice in tl. The total spectral widths covered 2000 Hz in FI and 6250 Hz in F2. In order to account for relaxation effects, the delay A was set to 5.0 ms, which is slightly shorter than l/(2JNH).

leaving the heteronuclear precession unaffected. Since proton chemical shifts are completely refocused, the experiment reveals pure absorptive signals in the F2 dimension. The signals in F,,owing to delayed t l acquisition, will be phase-modulated unless an additional refocusing pulse in the nitrogen channel is applied. A refocusing pulse coincident with one of the two proton mixing pulses restores the initial phase relations, thus allowing the nitrogen signals to be represented in a pure absorption mode. Since control of the signal phase is kept separate for each species of nucleus, the TPPI methodS is applicable for phase-sensitive detection of heteronuclear shifts. The experiment was performed on a Bruker AM-SO0 spectrometer equipped with a 5 mm reverse broad-band probe tuned to a 500.13 MHz proton frequency. A sample of fully ISN-labeled RNase T, was obtained from Escherichia coli by recombinant technique^.^^' The isoenzyme contains nine asparagine and two ~

( 5 ) Marion, D.; Wiithrich, 113, 967.

~~~~~

K. Biochem. Biophys. Res. Commun. 1983,

0 1990 American Chemical Society

J. Am. Chem. SOC.11990, 112, 1280-1281

1280

glutamine residues. Measurements were carried out with a protein concentration of 3 mM at pH 5.5 and 313 K. During relaxation, the solvent resonance was saturated by low-power excitation. The GARP decoupling scheme* was employed during acquisition. Approximately 4 W decoupling power at 50.68 MHz resulted in a i ~ / 2pulse length of 200 MS. The total acquisition time for the ISNHz-edited spectrum was 6 h, using an extended 64-step phase cycle. Figure 2A shows the spectral region of the selected amide correlations. A reference spectrum of all backbone and side-chain amide resonances (Figure 2B) was recorded according to the procedures of Bax e t al.9 An expansion (Figure 2C) reveals all 11 of the expected N H 2 correlations. In the arginine side chain, the protons undergo rapid exchange at the chosen temperature;'O therefore, they did not give rise to signals. Of the few proton resonances in RNase TInot yet assigned," the asparagine residues in positions 36, 43,44,98, and 99 are of particular interest as they are located close to the active site. Their N H 2 groups may be used as local spies for the investigation of base recognition and catalysis by means of NMR techniques. The addition of NOE or ROE transfer steps to the pulse sequence could eventually allow tracing of the connectivities to the carbon-bound protons of the amino acid side chains. The usefulness of this method for complete assignment of proton resonances in RNase T, is currently under investigation in our laboratory.

Table I. Literature Conclusions Regarding Regioselectivity of Protonation of the Radical Anion in the Birch Reduction of Anisole suggested preference meta meta ortho meta none meta ortho

basis qual mechanistic qual mechanistic Hiickel QM qual mechanistic ESR spin density SCF QM SCF OM

ref 19592 1959' 19614 1 9635 1 9666 1969' 1 9808

Scheme 1. Two Alternative Regioselectivities for the Reduction of Anisole OCH3

OCH3

ROHIOI

4 OCH~

80

90

OCH3

I

Acknowledgment. We thank Dr. R. Quaas and Dr. U. Hahn for providing the ISN-labeled RNase TI. Thanks are due the Deutsche Forschungsgemeinschaft for a grant. (6)Quaas. R.; McKeown, Y.;Stanssens, P.; Frank, R.; Bldcker, H.; Hahn, U.Eur. J . Biochem. 1988,173,611422. (7)Quaas, R.;Grunert, H.-P.; Kimura, M.; Hahn, U. Nucleosides Nucleotides 1988,7(5 and 6). 619-623. (8)Shaka, A.J.; Barker, P. B.; Freeman, R. J. Magn. Reson. 1985,64, 541-552. (9)Bax, A.; Griffey, R. H.; Hawkins, B. L. J . Magn. Reson. 1983,55, 301-3 15. (IO) Blomberg, F.; Maurer, W.; Riiterjans, H. Proc. Natl. Acad. Sci. U.S.A. 1976, 73(5), 1409-1413. (11) Hoffmann, E.;Riiterjans, H. Eur. J . Biochem. 1988,177,539-560.

Regioselectivity of the Birch Reduction Howard E. Zimmerman* and Patricia A. Wang Department of Chemistry, University of Wisconsin Madison, Wisconsin 53706 Received September 29, 1989 The Birch reduction' is one of the most fundamental reactions in organic chemistry. It is known2 that the reaction proceeds via (a) radical anion formation, (b) protonation by an alcohol, (c) addition of a second electron, and (d) protonation of the resulting cyclohexadienyl carbanion. In 1959 Krapcho and B0thner-B~~" provided evidence that the rate-limiting step is protonation of the radical anion.3b Note eq 1.

1

2

limiting

3

4

5

(1)

The regioselectivity follows the Birch rule that the 1,4-dihydrobenzene formed is the isomer having the maximum number ( I ) (a) Birch, A. J. J . Chem. SOC.1944,430. (b) Rabideau, P. W. Tetrahedron 1989,45, 1519. (2)Birch, A. J.; Nasipuri, D. Tetrahedron 1959,148. (3)(a) Krapcho, A. P.; Bothner-By, A. A. J. Am. Chem. Soc. 1959,81, 3658. (b) Reference 3a describes the kinetic third-order determination only for benzene; we therefore determined the order far anisole, which did prove to be third overall. This will be reported in a full paper.

0002-7863/90/1512-1280$02.50/0

of alkoxy1 and/or alkyl groups on the residual double bonds. One rationale for this orientation was presented nearly three decades ago! This suggested that there was preferential ortho protonation of the radical anion as the site of the highest electron density, based on Hiickel calculations. Anisole is an example that has been the object of particular debate. Whether the initial protonation is ortho or meta has been subject to an unusual variation of viewpoints and speculation as summarized in Table I. In this communication, we report the first experimental test of the initial protonation site in the Birch reduction of anisole. Our research began with and relies on the premise that the isotope selectivity in a protium-deuterium medium will be greater for a radical anion such as lthan a carbanion such as 9. Since radical anions are known to be considerably less basic and reactive species than carbanions: their protonation consequently will be less exothermic and more selective. Scheme I shows the two potential pathways for the reduction of anisole. We studied the Birch reduction of anisole in a sodium-liquid ammonia-tert-butyl alcohol medium enriched in deuterium. Due to kinetic isotope effects of our original premise, we anticipated that the product site with less deuterium content would be the carbon protonated in the radical anion protonation step. Conversely, the site with greater deuterium should be the one protonated as a nonselective carbanion reacting in a less selective process. The anisole reduction was run at -78 OC with sodium in liquid ammonia and tert-butyl alcohol containing ca. 2% deuterium. The ratio of m- to o-deuterium content in the 2,5-dihydroanisole product was determined by deuterium N M R analysis utilizing the Pr(fod), shift reagent to separate methylene absorption peaks. (4) (a) Zimmerman, H. E . Tetrahedron 1961,16, 169. (b) It was also noted that ortho protonation leads to a more delocalized radical. See: Zimmerman, H. E . In Molecular Rearrangements; de Mayo, P., Ed.; Interscience: New York, 1963;p 352. (5) Smith, H. Chemistry in Nonaqueous Ionizing Solvents; Interscience: New York, 1963;Vol. 1, Part 2, p 255. (6)Brown, J. K.; Burnham, D. R.; Rogers, N. A. J. Tetrahedron Lett. 1966,2621. (7)Burnham, D.R. Tetrahedron 1969,25, 897. (8)Birch, A.J.; Hinde, A. L.;Radom, L. J. Am. Chem. Soc. 1980,102, 3370. (9)Parker, V. D.;Tilset, M.; Hammerich, 0. J . Am. Chem. Sot. 1987, 109, 7905.

0 1990 American Chemical Society